Back to skill
Skillv1.0.0
ClawScan security
Taizi Alicloud Ai Image · ClawHub's context-aware review of the artifact, metadata, and declared behavior.
Scanner verdict
ReviewMar 16, 2026, 4:03 AM
- Verdict
- Review
- Confidence
- high
- Model
- gpt-5-mini
- Summary
- The skill appears to implement Qwen image generation correctly, but its runtime touches undeclared credentials and environment files (DASHSCOPE_API_KEY, ~/.alibabacloud/credentials, .env) which is inconsistent with the registry metadata and should be noted before installing.
- Guidance
- This skill appears to legitimately implement Alicloud (DashScope) image generation with Qwen models, but it will require and try to load a DASHSCOPE_API_KEY (from the DASHSCOPE_API_KEY env var or ~/.alibabacloud/credentials) and will also load .env files from the current directory and repo root. Before installing or running: 1) do not put sensitive keys into shared .env or workspace files unless you trust the skill; 2) verify the DASHSCOPE_API_KEY you provide has minimal permissions and is rotated if shared; 3) review the included scripts (they download images from URLs and will make network calls) and ensure that behavior is acceptable; and 4) note the registry metadata omission (no required env declared) — you should ensure the platform prompts for the API key or add it manually. Overall this looks functionally correct, but the undeclared credential access is a meaningful inconsistency.
Review Dimensions
- Purpose & Capability
- concernName/description match the code: it calls DashScope ImageGeneration with qwen-image models. However the skill metadata declares no required environment variables or primary credential, while both SKILL.md and the script require DASHSCOPE_API_KEY (or credentials in ~/.alibabacloud/credentials). That mismatch is disproportionate to the declared registry requirements.
- Instruction Scope
- concernRuntime instructions and the included script read environment variables, load .env files (from CWD and repo root found via .git), and load credentials from ~/.alibabacloud/credentials. The script also downloads reference images and generated image URLs via urllib. These behaviors are consistent with an image provider integration but the script will read local files that may contain secrets (unexpected given metadata) and will perform network fetches.
- Install Mechanism
- okThere is no automated install spec (instruction-only) which minimizes install-time risk. The SKILL.md recommends installing the dashscope Python package in a venv; the script will fail if dashscope isn't installed. No remote archives or obscure download URLs are used by the skill itself.
- Credentials
- concernThe skill requires DASHSCOPE_API_KEY and supports reading dashscope_api_key from ~/.alibabacloud/credentials and ALIBABA_CLOUD_PROFILE/ALICLOUD_PROFILE, but the registry lists no required env or primary credential. The number and type of env/config accesses are reasonable for an Alibaba Cloud SDK integration, but the omission from metadata is an inconsistency the user should be aware of.
- Persistence & Privilege
- okThe skill does not request persistent 'always' inclusion and does not attempt to modify other skills or system-wide agent settings. Its runtime activity (reading env/credentials, writing output images under workspace) is within expected scope for a provider integration.